Comparison of the power of the paired samples t test to that of Wilcoxon's signed-ranks test under various population shapes.

Used Monte Carlo methods to assess the relative power of the paired-samples t test and Wilcoxon's signed-ranks test (WSR) under 10 population shapes. Results indicate that each statistic was more powerful than the other in given situations; power advantages of the t under normal theory were small; in nonnormal situations, the t test never attained more than modest advantages. The WSR was more often the more powerful test, and the magnitude of the WSR's power advantage often increased with sample size. It is concluded that the claim that parametric tests are more powerful than nonparametric tests is not justified. (PsycINFO Database Record (c) 2010 APA, all rights reserved)